Orpo-Llama-3.2-1B-15k

AdamLucek/Orpo-Llama-3.2-1B-15k is an ORPO fine tuned version of meta-llama/Llama-3.2-1B on a subset of 15,000 shuffled entries of mlabonne/orpo-dpo-mix-40k.

Trained for 7 hours on an L4 GPU with this training script, modified from Maxime Labonne's original guide

For full model details, refer to the base model page meta-llama/Llama-3.2-1B

Evaluations

Benchmark	Accuracy	Notes
AGIEval	20.99%	Average across multiple reasoning tasks
GPT4ALL	51.12%	Average across all categories
TruthfulQA	42.80%	MC2 accuracy
BigBench	31.75%	Average across 18 tasks
MMLU	31.23%	Average across all categories
Winogrande	61.33%	5-shot evaluation
ARC Challenge	35.92%	25-shot evaluation
HellaSwag	48.65%	10-shot evaluation

Detailed Eval Metrics Available Here

Using this Model

from transformers import AutoTokenizer
import transformers
import torch

# Load Model and Pipeline
model = "AdamLucek/Orpo-Llama-3.2-1B-15k"

pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

# Load Tokenizer
tokenizer = AutoTokenizer.from_pretrained(model)

# Generate Message
messages = [{"role": "user", "content": "What is a language model?"}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

Training Statistics